Underwater image restoration using level adjustment and colour equalization

Abstract

Due to the light absorption and scattering by water, the underwater image quality is often severely degraded. To solve this problem, we propose an underwater image restoration method that uses level adjustment and colour equalization algorithms to obtain high-quality underwater images. A human visual perception-based underwater imaging model is first built to generate images with high contrast and rich colours. Then, a level adjustment algorithm is developed to improve the brightness of the image. Finally, a colour equalization algorithm is designed to filter the local characteristics and achieve image brightness, colour, and contrast adjustment with nonlinear and filtered local characteristics. Extensive numerical and experimental results show that the proposed method can improve the underwater images and is superior to the recently reported methods.